Alibaba has launched its new Qwen3.8-Max large language model, boasting a 2.4 trillion parameter count and enhanced capabilities in coding and professional office tasks. This release has boosted Alibaba's stock, with the model's API now available on the Qwen AI platform and integrated into a new agent product called "Qianwen Office." In parallel, DeepSeek's V4 Flash model has seen a massive surge in usage, processing 8 trillion tokens in a single day and topping usage charts, prompting OpenAI to significantly reduce prices for its GPT-5.6 Luna and Terra models in response to the competitive pressure. AI
IMPACT Intensifies competition in the LLM market, driving down API costs and pushing innovation in agent capabilities.
RANK_REASON Frontier-lab model release with system card and competitive pricing response. [lever_c_demoted from frontier_release: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
